BACKGROUNDWith each successive semiconductor technology generation, wafer diameters tend to increase and transistor sizes decrease, resulting in the need for an ever higher degree of accuracy and repeatability in substrate processing. Semiconductor substrate materials, such as silicon wafers, are routinely processed using plasma in vacuum chambers. Plasma processing techniques include sputter deposition, plasma-enhanced chemical vapor deposition (PECVD), resist strip, and plasma etch.
In a plasma processing chamber, process gas is excited into a plasma in a proximity of a substrate being processed. However, gas at other locations (such as gas holes, conduits, lift pin holes, plenums, and the like) in the plasma processing chamber can also be excited into an unwanted plasma under certain conditions. Such an unwanted plasma is called parasitic plasma. Parasitic plasma can occur in a plasma processing chamber due to many reasons, such as geometry of chamber components, gas pressure and gas chemistry used in plasma processing recipes, supply of radio frequency (RF) power, etc. Parasitic plasma can cause a variety of issues in plasma processing, such as particle contamination, spatial and temporal nonuniformity during processing, and/or premature failure of chamber components. It is therefore desirable to eliminate parasitic plasma in a plasma processing chamber.
SUMMARYDescribed herein is a component of a plasma processing chamber, comprising: a first layer of a non-metal material, the first layer having first and second opposing surfaces, the first surface exposed to plasma in the plasma processing chamber and the second surface not exposed plasma in the plasma processing chamber; a second layer of an electrically conductive material, the second layer bonded to the second surface of the first layer; a void space extending through the entire first and second layers; a tubular sleeve lining an interior surface of the void space in the second layer such that the second layer is not exposed in the void space and parasitic plasma can be prevented in the void space; wherein one end of the tubular sleeve is coplanar with facing surfaces of the first layer and the second layer and the other end of the tubular sleeve is coplanar with a lower surface of the second layer.
BRIEF DESCRIPTION OF FIGURESFIG. 1 shows a schematic of an exemplary plasma processing chamber.
FIG. 2 shows a cross sectional view of a component of a plasma processing chamber, the component having a void space.
FIG. 3 shows a cross sectional view of a component of a plasma processing chamber, the component having a void space with a sleeve lining an interior surface thereof, according to a first embodiment.
FIG. 4 shows a cross sectional view of a component of a plasma processing chamber, the component having a void space with a sleeve lining an interior surface thereof, according to a second embodiment.
FIG. 5 shows a cross sectional view of a component of a plasma processing chamber, the component having a void space with a sleeve lining an interior surface thereof, according to a third embodiment.
DETAILED DESCRIPTIONDescribed herein is a method and components of a plasma processing chamber for eliminating parasitic plasma. The method and components are not limited to a particular type of plasma processing chamber or a particular plasma processing technique. A plasma processing chamber can rely on a variety of mechanisms to generate plasma, such as inductive coupling (transformer coupling), helicon, electron cyclotron resonance, capacitive coupling (parallel plate). For instance, high density plasma can be produced in a transformer coupled plasma (TCP™) processing chamber, or in an electron cyclotron resonance (ECR) processing chamber. Transformer coupled plasma processing chambers, wherein RF energy is inductively coupled into the chambers, are available from Lam Research Corporation, Fremont, Calif. An example of a high-flow plasma processing chamber that can provide high density plasma is disclosed in commonly-owned U.S. Pat. No. 5,948,704, the disclosure of which is hereby incorporated by reference. Parallel plate plasma processing chambers, electron-cyclotron resonance (ECR) plasma processing chambers, and transformer coupled plasma (TCP™) processing chambers are disclosed in commonly-owned U.S. Pat. Nos. 4,340,462; 4,948,458; 5,200,232 and 5,820,723, the disclosures of which are hereby incorporated by reference.
By way of example, plasma can be produced in a parallel plate processing chamber such as the dual frequency plasma etching chamber described in commonly-owned U.S. Pat. No. 6,090,304, the disclosure of which is hereby incorporated by reference. A preferred parallel plate plasma processing chamber is a dual frequency capacitively coupled plasma processing chamber including an upper showerhead electrode and a substrate support. For purposes of illustration, embodiments herein are described with reference to a parallel plate type plasma processing chamber.
A parallel plate plasma processing chamber for plasma etching is illustrated inFIG. 1. Theplasma processing chamber100 comprises achamber110, aninlet load lock112, and an optionaloutlet load lock114, further details of which are described in commonly-owned U.S. Pat. No. 6,824,627, which is hereby incorporated by reference in its entirety.
Theload locks112 and114 (if provided) include transfer devices to transfer substrates such as wafers from awafer supply162, through thechamber110, and out to awafer receptacle164. Aload lock pump176 can provide a desired vacuum pressure in theload locks112 and114.
Avacuum pump172 such as a turbo pump is adapted to maintain a desired pressure in thechamber110. During plasma etching, the chamber pressure is controlled, and preferably maintained at a level sufficient to sustain a plasma. Too high a chamber pressure can disadvantageously contribute to etch stop while too low a chamber pressure can lead to plasma extinguishment. In a medium density plasma processing chamber, such as a parallel plate plasma processing chamber, preferably the chamber pressure is maintained at a pressure below about 200 mTorr (e.g., less than 100 mTorr such as 20 to 50 mTorr) (“about” as used herein means±10%).
Thevacuum pump172 can be connected to an outlet in a wall of thechamber110 and can be throttled by avalve173 in order to control the pressure in the chamber. Preferably, the vacuum pump is capable of maintaining a pressure within thechamber110 of less than 200 mTorr while etching gases are flowed into thechamber110.
Thechamber110 includes anupper electrode assembly120 including an upper electrode125 (e.g., showerhead electrode), and asubstrate support150. Theupper electrode assembly120 is mounted in anupper housing130. Theupper housing130 can be moved vertically by amechanism132 to adjust the gap between theupper electrode125 and thesubstrate support150.
Aprocess gas source170 can be connected to thehousing130 to deliver process gas comprising one or more gases to theupper electrode assembly120. In a preferred plasma processing chamber, the upper electrode assembly comprises a gas distribution system, which can be used to deliver process gas to a region proximate to the surface of a substrate. Gas distribution systems, which can comprise one or more gas rings, injectors and/or showerheads (e.g., showerhead electrodes), are disclosed in commonly-owned U.S. Pat. Nos. 6,333,272; 6,230,651; 6,013,155 and 5,824,605, the disclosures of which are hereby incorporated by reference.
Theupper electrode125 preferably comprises a showerhead electrode, which includes gas holes (not shown) to distribute process gas therethrough. The gas holes can have a diameter of 0.02 to 0.2 inch. The showerhead electrode can comprise one or more vertically spaced-apart baffle plates that can promote the desired distribution of process gas. The upper electrode and the substrate support may be formed of any suitable material such as graphite, silicon, silicon carbide, aluminum (e.g., anodized aluminum), or combinations thereof. A heattransfer liquid source174 can be connected to theupper electrode assembly120 and another heat transfer liquid source can be connected to thesubstrate support150.
Thesubstrate support150 can have one or more embedded clamping electrodes for electrostatically clamping a substrate on an upper surface155 (support surface) of thesubstrate support150. Thesubstrate support150 can be powered by an RF source and attendant circuitry (not shown) such as RF matching circuitry. Thesubstrate support150 is preferably temperature controlled and may optionally include a heating arrangement (not shown). Examples of heating arrangements are disclosed in commonly assigned U.S. Pat. Nos. 6,847,014 and 7,161,121, which are hereby incorporated by reference. Thesubstrate support150 can support a semiconductor substrate such as a flat panel or 200 mm or 300 mm wafer on thesupport surface155.
Thesubstrate support150 preferably includes passages therein for supplying a heat transfer gas such as helium under the substrate supported on thesupport surface155 to control the substrate temperature during plasma processing thereof. For example, helium back cooling can maintain wafer temperature low enough to prevent burning of photoresist on the substrate. A method of controlling a temperature of a substrate by introducing a pressurized gas into a space between the substrate and the substrate support surface is disclosed in commonly-owned U.S. Pat. No. 6,140,612, the disclosure of which is hereby incorporated by reference.
Thesubstrate support150 can include lift pin holes (not shown), through which lift pins can be actuated vertically by suitable mechanisms and raise the substrate off thesupport surface155 for transport into and out from thechamber110. The lift pin holes can have a diameter of about 0.08 inch. Details of lift pin holes are disclosed in commonly owned U.S. Pat. Nos. 5,885,423 and 5,796,066, the disclosure of which is hereby incorporated by reference.
Void spaces such as the gas holes, helium passages, lift pin holes in theupper electrode125 and thesubstrate support150 can be prone to parasitic plasma. For example, if a void space extends across layers of materials with different relative permittivity at RF frequencies, excessive RF voltage can occur between the layers, under certain conditions (e.g. gas pressure, RF load, RF frequency, etc.). Such excessive RF voltage (e.g. 20 V or more) can be sufficient to cause parasitic plasma in the void space. Electrically conductive materials commonly used in theupper electrode125 and thesubstrate support150 can include aluminum, steel, graphite, and doped silicon. Exposed conductive interior surfaces of the void space (e.g. of gas holes) can concentrate electric fields and intensify parasitic plasma in their proximity, when RF power is supplied to theprocessing chamber100 to generate plasma therein. DC breakdown voltage VBof a gas in parallel plates as a function of pressure p and gap distance d is given by the Paschen
where A and B are constants determined by the properties of the gas (e.g., temperature, gas composition and ionization potential) and γ is a parameter related to materials of the parallel plates. AC breakdown voltage can be as little as one tenth of VB, especially in the vicinity of conductive surfaces that are exposed to the gas and concentrated electric fields.
FIG. 2 shows a cross sectional diagram of avoid space210 in a component200 (e.g. an upper electrode or a substrate support) of a plasma processing chamber. Thecomponent200 comprises alayer220 of a non-metal material (e.g. ceramic or undoped silicon), alayer230 of an electrically conductive material (e.g. aluminum). Thelayer220 has asurface220aexposed toplasma260 in the plasma processing chamber and asurface220bnot exposed toplasma260. Thelayer230 can be bonded to thesurface220busing a suitable technique such as using an adhesive (e.g. silicone rubber). Eachvoid space210 extends through theentire layer220 and theentire layer230 and is open to an interior of the plasma processing chamber. The thickness of thelayer220 and the thickness of thelayer230 are preferably from 0.5 to 5 mm respectively. Thevoid space210 can be about 0.02 to about 0.08 inch in diameter. Thelayer220 can be made of, for example, alumina, aluminum nitride, undoped silicon, silicon nitride, silicon oxide, silicon oxynitride and/or yttria. Thelayer230 can be made of, for example, metal, graphite and/or doped silicon. Thelayer230 has aconductive surface235 exposed in an interior of thevoid space210.Parasitic plasma250 can occur in the interior of thevoid space210 and cause erosion and/or corrosion therein. Thecomponent200 can be attached (such as bonded, supported, fastened and/or adhered) to abase290. The base can be temperature controlled, for example, by flowing liquid through embeddedfluid channels290atherein. The base can also have at least oneplenum290bin fluid communication with thevoid210. RF power can be supplied to the base290 to generate theplasma260. Thelayer220 can have at least one electrostatic chuck (ESC)electrode220cembedded therein. At least one heater240 (e.g. thin film heater) is preferably in thermal contact with (e.g. attached to a bottom surface of or embedded in) thelayer230. Preferably, a relative permittivity of the electrical conductive material oflayer230 at radio frequencies present in the plasma processing chamber is at least 20 times greater than a relative permittivity of the non-metal material oflayer220 at the radio frequencies. In an example, wherein thevoid space210 is filled with helium gas at 30 Torr; thelayer220 is approximately 0.1 inch thick; thelayer230 is bonded to thelayer220 with an adhesive layer of approximately 0.03 inch thick; the component is subject to RF power of 2500 W at 27 MHz; and theplasma260 has an impedance of approximately 60Ω, an RF potential of approximately 15.5 V will develop between thelayer230 and thesurface220a, which can exceed the breakdown voltage of the helium gas in thevoid space210 and causeparasitic plasma250 therein.
In one embodiment, as shown inFIG. 3, atubular sleeve300 lines an interior surface of thevoid space210. Thesleeve300 is preferably made of a material with a dielectric constant lower than that of thelayer230, such as plastic (e.g. polyether ether ketone) or ceramic (e.g. alumina). Preferably, the portion of thevoid space210 inlayer220 has the same cross sectional shape (e.g. circle, polygon or any other suitable shape) as the passage through thesleeve300. Thevoid space210 can be in the shape of a cylinder or a prism. Specifically, if thevoid space210 is cylindrical in shape, thetubular sleeve300 has an inner diameter equal to an inner diameter of thevoid space210 in thelayer220. The radial thickness of thesleeve300 should be sufficient to suppress parasitic plasma in thevoid space210, preferably at least 0.01 inch. Preferably, thesleeve300 is made of the same material as thelayer220 and is not directly exposed to theplasma260. Thesleeve300 preferably has one end coplanar with an upper surface oflayer230 and the other end of thesleeve300 is preferably coplanar with a lower surface oflayer230 such that theconductive surface235 of thelayer230 is entirely covered by thesleeve300, i.e. no conductive surface is exposed in thevoid space210. Preferably, thesleeve300 does not extend into thebase290.
Thesleeve300 can be a self-supporting sleeve affixed to thecomponent200 by any suitable means, such as adhesive, press fitting or the like. However, thesleeve300 can have threads on its exterior surface that mate with threads in thelaminate200. Thesleeve300 can also be in a form of a coating applied on the interior surface of thevoid space210 by any suitable coating means. For example, CVD, plasma spraying.
In another embodiment, as shown inFIG. 4, atubular sleeve400 is the same as thesleeve300 except that thesleeve400 has anupper flange410. Theflange410 preferably has an upper surface coplanar with an upper surface oflayer230. Thesleeve400 can be mounted in thevoid space210 before bonding thelayer220 tolayer230.
FIG. 5 shows yet another embodiment. Atubular sleeve500 is the same as thesleeve300 except that thesleeve500 has alower flange510. In this case, the flange has a lower surface coplanar with a lower surface oflayer230. Thesleeve500 can be mounted in thevoid space210 before attaching the base290 to thelayer230.
The sleeve described herein can also be used in cavities, holes, conduits, voids, plenums and/or other spaces prone to parasitic plasma in a plasma processing chamber. The sleeve can be made into a shape that fits interior conductive surfaces of these spaces. For example, thesubstrate support150 can have sleeves in lift pin holes and/or helium passages.
While the sleeve for eliminating parasitic plasma, and the plasma exposed laminate having the sleeve therein have been described in detail with reference to specific embodiments thereof, it will be apparent to those skilled in the art that various changes and modifications can be made, and equivalents employed, without departing from the scope of the appended claims.